INDEX

Explanations

mentions of document categories or classifications

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æŁĦ

-0.19

igor

-0.17

hea

-0.17

abilia

-0.17

BoundingBox

-0.16

cura

-0.15

á»Ļc

-0.14

_qp

-0.14

aid

-0.14

 verst

-0.14

POSITIVE LOGITS

0.17

0.15

fro

0.15

ords

0.15

ogg

0.14

Rag

0.14

Peg

0.14

 suit

0.14

atics

0.14

 Under

0.14

Activations Density 0.004%

mentions of document categories or classifications

No Comments

No Known Activations